https://github.com/cran/spatstat
Raw File
Tip revision: 63c82df34a6e9f19fa18a5466038e9e660e8283e authored by Adrian Baddeley on 14 May 2008, 00:00:00 UTC
version 1.13-2
Tip revision: 63c82df
as.im.R
#
#    as.im.R
#
#    conversion to class "im"
#
#    $Revision: 1.12 $   $Date: 2007/11/11 13:56:27 $
#
#    as.im()
#
as.im <- function(X, W=as.mask(as.owin(X), dimyx=dimyx), ...,
                  dimyx=NULL, na.replace=NULL) {

  na.handle <- function(X, na.replace) {
    if(is.null(na.replace))
      return(X)
    if(length(na.replace) != 1)
      stop("na.replace should be a single value")
    X$v[is.na(X$v)] <- na.replace
    return(X)
  }
  
  if(verifyclass(X, "im", fatal=FALSE)) {
    if(missing(W) && is.null(dimyx))
      return(na.handle(X, na.replace))

    # reshape pixel raster
    # invoke W = as.mask(X, dimyx)
    Y <- as.im(W, dimyx=dimyx)
    phase <- c((Y$xcol[1] - X$xcol[1])/X$xstep,
               (Y$yrow[1] - X$yrow[1])/X$ystep)
    Y$v <- matrixsample(X$v, Y$dim, phase=round(phase))
    return(na.handle(Y, na.replace))
  }

  if(verifyclass(X, "owin", fatal=FALSE)) {
    # if W is missing, the default is now evaluated, as above.
    # if W is present, it may have to be converted
    if(!missing(W)) {
      stopifnot(is.owin(W))
      if(W$type != "mask")
        W <- as.mask(W, dimyx=dimyx)
    }
    m <- W$m
    v <- m * 1
    v[!m] <- NA
    out <- list(v = v, 
                dim    = W$dim,
                xrange = W$xrange,
                yrange = W$yrange,
                xstep  = W$xstep,
                ystep  = W$ystep,
                xcol   = W$xcol,
                yrow   = W$yrow,
                lev    = NULL,
                type    = "integer",
                units  = unitname(X))
    class(out) <- "im"
    return(na.handle(out, na.replace))
  }

  if((is.vector(X) || is.factor(X)) && length(X) == 1) {
    xvalue <- X
    X <- function(xx, yy, ...) { rep(xvalue, length(xx)) }
  }
  
  if(is.function(X)) {
    f <- X
    W <- as.owin(W)
    W <- as.mask(W, dimyx=dimyx)
    m <- W$m
    funnywindow <- !all(m)

    xx <- as.vector(raster.x(W))
    yy <- as.vector(raster.y(W))
    lev <- NULL

    # evaluate function value at each pixel 
    if(!funnywindow) 
      values <- f(xx, yy, ...)
    else {
      # evaluate only inside window
      inside <- as.vector(m)
      val <- f(xx[inside], yy[inside], ...)
      # create space for full matrix
      msize <- length(m)
      values <-
        if(!is.factor(val))
          vector(mode=typeof(val), length=msize)
        else {
          lev <- levels(val)
          factor(rep(lev[1], msize), levels=lev)
        }
      # copy values, assigning NA outside window
      values[inside] <- val
      values[!inside] <- NA
    }

    if(is.factor(values)) 
        lev <- levels(values)

    out <- im(values, W$xcol, W$yrow, lev, unitname=unitname(W))
    return(na.handle(out, na.replace))
  }

  if(is.list(X) && checkfields(X, c("x","y","z"))) {
    stopifnot(is.matrix(X$z))
    z <- X$z
    y <- X$y
    x <- X$x
    # Usual S convention as in contour.default() and image.default()
    # Rows of z correspond to x values.
    nr <- nrow(z)
    nc <- ncol(z)
    lx <- length(x)
    ly <- length(y)
    if(lx == nr + 1)
      x <- (x[-1] + x[-lx])/2
    else if(lx != nr)
      stop("length of x coordinate vector does not match number of rows of z")
    if(ly == nc + 1)
      y <- (y[-1] + y[-ly])/2
    else if(ly != nc)
      stop("length of y coordinate vector does not match number of columns of z")
    # convert to class "im"
    out <- im(t(z), x, y)
    # now apply W and dimyx if present
    if(missing(W) && !is.null(dimyx))
      out <- as.im(out, dimyx=dimyx)
    else if(!missing(W))
      out <- as.im(out, W=W, dimyx=dimyx)
    return(na.handle(out, na.replace))
  }
    
  stop("Can't convert X to a pixel image")
}
back to top